The goal of this project is to understand the molecular basis of virus persistence within the central nervous system (CNS). Persistent viral infections are associated with a number of chronic human CNS infections, including human immunodeficiency virus and measles virus. This project uses the JHM strain of mouse hepatitis virus (JHMV) as a model system. JHMV produces an acute encephalitis with demyelination and chronic infection of the murine CNS as demonstrated by persistent antigen, and ongoing primary demyelination. The pathological hallmarks of this chronic disease are similar to the demyelination associated with multiple sclerosis. This project targets two potential mechanisms of virus persistence. Our first hypothesis is that persistence is due to the inability of the immune system to effectively clear virus from the CNS. Analysis of this mechanism is based on the recently defined cytotoxic T lymphocyte (CTL) epitopes found within two virus proteins. Single amino acid changes within these epitopes will be tested for their ability to prevent target recognition or antagonize these critical effectors as a mechanism for the generation of CTL escape mutants, resulting in viral persistence. This will be accomplished by testing the responses of CTL clones to peptides containing single amino acid substitutions within the optimal epitopes. Sequence analysis of the viral genes encoding both epitopes will be analyzed during chronic disease to identify mutations associated with viral persistence. Epitopes containing relevant mutations will be expressed in the CNS during infection using a novel defective interfering (DI) vector system. This vector requires helper virus for replication and gene expression, thus altered epitopes will only be expressed within the CNS. Our second hypothesis is that persistence is associated with mutation in viral genes encoding regulatory elements or genes required for virus assembly. This mechanism is based on the presence of viral antigen but absence of infectious virus in CNS of chronically infected mice. Viral antigen in the absence of infectious virus suggests defects in either the genes regulating viral replication or assembly. Sequences of specific regions, with either regulatory or assembly functions will be analyzed in a focused approach to define mutations affecting these functions. For each potential gene product assays are described to directly test the relevance of these mutation to replication or assembly. This project provides a unique focused approach to the understanding of virus persistence in general with a special emphasis on the ability of virus to persist within the CNS.